Together · Blogs

Blogs listings from Together on TechiSeek. 208 public listings.

• AI for Systems: Using LLMs to Optimize Database Query Execution (Together)
  We worked in collaboration with Stanford University, the University of Wisconsin–Madison, and Bauplan to test whether LLMs can optimize dat…
• Mamba-3B-SlimPJ: State-space models rivaling the best Transformer architecture (Together)
  The Mamba architecture, building on a long line of work on state-spaces models (e.g S4) and hardware-efficient algorithms (e.g.
• Together AI and Meta partner to bring PyTorch Reinforcement Learning to the AI Native Cloud (Together)
  Together AI and the Py Torch team at Meta are partnering to build an open-source reinforcement learning framework for training agentic AI s…
• Together Link: open models in the harness you already use. Start with one command today. (Together)
  Together Link connects the harness your team already uses to the best open models on Together AI and cuts your spend by over 50%.
• New in Together GPU Clusters: Reliability and control for production GPU clusters (Together)
  The first is platform health: passive health checks, auto node repair, and Slinky 1.0, focused on catching and recovering from the failure…
• Together AI and Y Combinator partner to launch the first dedicated GPU cluster for the YC community (Together)
  Today, Together AI and Y Combinator (YC) are announcing a partnership to deliver the first dedicated YC GPU cluster, giving YC's portfolio…
• Build real-time voice agents on Together AI (Together)
  Call (847) 851-4323 and ask our assistant about models, pricing, or how to deploy a voice agent.
• Model-Preserving Adaptive Rounding with YAQA (Together)
  We’re excited to announce YAQA (Yet Another Quantization Algorithm), a new weight-only LLM post-training quantization method that quantizes…
• Announcing Together Inference Engine (Together)
  The Together Inference Engine is multiple times faster than any other inference service, with 117 tokens per second on Llama-2-70 B-Chat an…
• How to run TorchForge reinforcement learning pipelines in the Together AI Native Cloud (Together)
  Torch Forge RL pipelines now run on Together AI’s Instant Clusters with support for distributed training and sandboxed environments.
• Together AI partners with Meta to release Llama 3.1 models for inference and fine-tuning with (Together)
  Today marks an inflection point for open source AI with the launch of Meta Llama 3.1 405 B, the largest openly available foundation model,…
• Together AI acquires CodeSandbox to launch first-of-its-kind code interpreter for generative AI (Together)
  Code generation has quickly become one of the most popular use cases for LLMs. However, while LLMs excel at generating code, they cannot ex…
• Multimodal Document RAG with Llama 3.2 Vision and ColQwen2 (Together)
  For AI models to be effective in specialized tasks, they often require domain-specific knowledge.
• Together AI Brings NVIDIA Nemotron 3 to Developers on Day 0 (Together)
  We’re excited to bring NVIDIA Nemotron 3 Super to Together AI, the AI Native Cloud. Built for multi-agent orchestration and complex reasoni…
• OpenChatKit now runs on consumer GPUs with a new 7B parameter model (Together)
  Open Chat Kit provides a powerful, open-source base to create both specialized and general purpose chatbots.
• Deploy Leading AI Models Accelerated by NVIDIA NIM on Together AI (Together)
  Together AI is a leader at scaling the deployment of generative AI models with fast performance and industry-leading cost efficiency.
• Together Code Sandbox: the most robust infrastructure for building AI coding products at scale (Together)
  The world runs on code, and most of that code is now being generated by LLMs. Powered by a robust infrastructure, Together Code Sandbox pro…
• Canary rollouts: upgrade models in production without downtime (Together)
  Rollouts move live traffic from your current model to a new checkpoint in gated steps. Health checks always run before traffic moves; on a…
• Optimizing Training Workloads for GPU Clusters (Together)
  This article outlines best practices focused on optimizing training workloads on GPU clusters.
• NeurIPS 2022: Overcoming communication bottlenecks for decentralized training (1/2) (Together)
  At Together we are working to bring the world’s computation together in form of a decentralized cloud, to enable researchers and practition…
• Parcae: Doing more with fewer parameters using stable looped models (Together)
  We present Parcae, one of the first stable architectures for looped language models, achieving the quality of a Transformer twice the size…
• Decentralized training of foundation models in heterogeneous environments (Together)
  Training foundation models, such as GPT-3 and PaLM, can be extremely expensive, often involving tens of thousands of GPUs running continuou…
• Qwen3-Coder: The Most Capable Agentic Coding Model Now Available on Together AI (Together)
  Code Smarter with Qwen3-Coder on Together AI's frontier AI cloud Starting today on Together AI, you can access Qwen3-Coder-480 B-A35 B-Inst…
• Can you feel the MoE? Mixtral available with over 100 tokens per second through Together Platform! (Together)
  Today, Mistral released Mixtral 8 x7 B, a high-quality sparse mixture of experts model (SMoE) with open weights.
• Chipmunk: Training-Free Acceleration of Diffusion Transformers with Dynamic Column-Sparse Deltas (Together)
  TL;DR: We present Chipmunk, a training-free method to accelerate diffusion transformers with hardware-aware dynamic sparsity.
• Together AI expands fine-tuning service with more models, live metrics, and finer controls (Together)
  Turning an open-weight model into a high-performing model for your task takes a sequence of well-measured experiments.
• Dynamic AI agent testing for the real world with Collinear Simulations and Together Evals (Together)
  Build measurably robust AI agents for real-world scenarios with Collinear Simulations and Together Evals API.
• Together AI partners with Meta to release Meta Llama 3 for inference and fine-tuning (Together)
  Together AI is proud be a launch partner for Meta Llama 3 on the new Together Inference Engine providing best in class performance up to 35…
• Optimizing inference speed and costs: Lessons learned from large-scale deployments (Together)
  How can teams reduce inference latency without massive costs? Achieving faster inference doesn't always mean paying more for a bigger clust…
• Introducing Together Code Sandbox & Together Code Interpreter: SOTA code execution for AI (Together)
  LLMs have revolutionized the way we generate code, but running that code in a secure, customizable, and performant environment remains a ch…
• Together AI brings Thinking Machines Lab’s new model Inkling on day 0 (Together)
  Today, Thinking Machines Lab released Inkling, a new multimodal mixture-of-experts model built for token-efficient reasoning, native multim…
• Announcing OpenChatKit (Together)
  Open Chat Kit provides a powerful, open-source base to create both specialized and general purpose chatbots for various applications.
• To Infinity and Beyond: ThunderKittens Now on NVIDIA Vera Rubin NVL72! (Together)
  The kernels team at Together recently received access to the NVIDIA Vera Rubin NVL72 platform.
• Announcing General Availability of Together Instant Clusters, offering ready to use, self-service (Together)
  Standing up multi‑node GPU clusters has historically been too manual and brittle — tickets, contract steps, and complex setup – taking valu…
• Inside multi-node training: How to scale model training across GPU clusters (Together)
  Training foundation models requires orchestrating hundreds or thousands of GPUs working in parallel.
• HELM: benchmarking large language models on the Together Research Computer (Together)
  Together’s software network harnessed spare GPU cycles across thousands of servers to benchmark 10 prominent open language models and proce…
• Announcing Together Inference Engine 2.0 with new Turbo and Lite endpoints (Together)
  The Together Inference Engine achieves over 400 tokens per second on Meta Llama 3 8 B by building on the latest advances from Together AI,…
• Together AI acquires Refuel.ai to unlock data for developers and businesses building (Together)
  If you speak with the largest enterprises today, they will tell you about the hundreds of AI applications they would like to deploy across…
• Continued Fine-tuning of LLMs: A Technical Deep Dive (Together)
  Keeping large language models updated with new knowledge and skills relevant to businesses remains challenging.
• Salesforce, Zoom, InVideo Train Faster with Together AI Turbocharged with NVIDIA Blackwell (Together)
  At NVIDIA GTC, we shared our bold plans to scale out thousands of NVIDIA Blackwell GPUs as an NVIDIA Cloud Partner, launch self-serve Insta…
• Together AI Announces $305M Series B to Scale AI Acceleration Cloud for Open Source and Enterprise (Together)
  Today marks an exciting milestone in Together AI's journey. We're thrilled to announce our $305 million Series B funding round, led by Gene…
• Together AI welcomes Kai Mak as CRO to accelerate gen AI adoption for AI natives and enterprises (Together)
  At Together AI, we believe the future of Generative AI hinges on open research, open science, and trust among researchers, developers, and…
• Together AI and NVIDIA collaborate to power Llama 3.1 models for enterprises on NVIDIA DGX Cloud (Together)
  We are thrilled to announce our collaboration with NVIDIA that brings the industry-leading Together Inference Engine to NVIDIA AI Foundry c…
• BASED: Simple linear attention language models balance the recall-throughput tradeoff (Together)
  Introducing Based, a simple efficient architecture that combines two familiar primitives – sliding window attention and linear attention –…
• Benchmarking inference at scale: coding agents (Together)
  On a production coding agent workload, Together Inference Engine delivers 31% more TPS than the next fastest OSS engine on the same hardwar…
• Together AI Brings NVIDIA Nemotron 3 Nano Omni to Developers on Day 0 (Together)
  NVIDIA Nemotron™ 3 Nano Omni is now available on the Together AI platform. Representing a meaningful step forward for multimodal AI, Nemotr…
• Cache-aware prefill–decode disaggregation (CPD) for up to 40% faster long-context LLM serving (Together)
  Serving long prompts doesn’t have to mean slow responses. At Together AI, we built cache-aware prefill–decode disaggregation (CPD), a servi…
• CocktailSGD: Fine-tuning foundation models over 500Mbps networks (Together)
  Distributed training of foundation models, especially large language models (LLMs), is communication-intensive and so has heavily relied on…
• Medusa: Simple framework for accelerating LLM generation with multiple decoding heads (Together)
  Large Language Models (LLMs) have changed the world. However, generating text with them can be slow and expensive.
• Announcing Together Python SDK v2.0 (Together)
  - 📚 Migration guide: We’ve published a detailed Python SDK Migration Guide covering API‑by‑API changes, type updates, and troubleshooting t…